linux-kernel - Re: [PATCH 3/3] status: Display an informational message when the VSTATUS character is pressed or TIOCSTAT ioctl is called.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YeamX3ELWyc/BtGp@hirez.programming.kicks-ass.net>
Date:   Tue, 18 Jan 2022 12:37:03 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Jiri Slaby <jirislaby@...nel.org>
Cc:     Walt Drummond <walt@...mmond.us>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>, ar@...msu.ru,
        linux-kernel@...r.kernel.org, Oleg Nesterov <oleg@...hat.com>,
        mm <linux-mm@...ck.org>, Ingo Molnar <mingo@...hat.com>
Subject: Re: [PATCH 3/3] status: Display an informational message when the
 VSTATUS character is pressed or TIOCSTAT ioctl is called.

On Tue, Jan 18, 2022 at 11:29:54AM +0100, Jiri Slaby wrote:

> > +static const char * const task_state_array[] = {
> > +	"running",
> > +	"sleeping",
> > +	"disk sleep",
> > +	"stopped",
> > +	"tracing stop",
> > +	"dead",
> > +	"zombie",
> > +	"parked",
> > +	"idle",
> > +};

Please no; don't add yet another one of these things. Can't you use the
one in fs/proc/array.c ?


> > +static inline struct task_struct *compare(struct task_struct *new,
> > +					  struct task_struct *old)
> > +{
> > +	unsigned int ostate, nstate;
> > +
> > +	if (old == NULL)
> > +		return new;
> > +
> > +	ostate = task_state_index(old);
> > +	nstate = task_state_index(new);
> > +
> > +	if (ostate == nstate) {

That's not an ordered set, please don't do that.

> > +		if (old->start_time > new->start_time)
> > +			return old;
> > +		return new;
> > +	}
> > +
> > +	if (ostate < nstate)
> > +		return old;
> > +
> > +	return new;
> > +}

> > +static inline const char *get_task_state_name(struct task_struct *tsk)
> 
> This definitely doesn't belong here. How do you ensure it matches the
> returned index also in the future. Put it along with task_index_to_char()?
> Or simply use task_state_to_char()?
> 
> > +{
> > +
> > +	int index;
> > +
> > +	index = task_state_index(tsk);
> > +	if (index > ARRAY_SIZE(task_state_array))
> 
> Should be >=, or?
> 
> > +		return task_state_unknown;
> > +	return task_state_array[index];
> > +}

*groan*.. that's very bad copy paste from fs/proc/array.c. There at east
there's a BUILD_BUG_ON() to make sure things are good.

> > +
> > +int n_tty_get_status(struct tty_struct *tty, char *msg, size_t *msglen)
> > +{
> > +	unsigned long loadavg[3];
> > +	uint64_t pcpu, cputime, wallclock;
> > +	struct task_struct *p;
> > +	struct rusage rusage;
> > +	struct timespec64 utime, stime, rtime;
> > +	char tname[TASK_COMM_LEN];
> 
> How much stack did you consume in sum with its caller n_tty_status()?
> 
> > +	size_t len;
> > +
> > +	if (tty == NULL)
> > +		return -ENOTTY;
> 
> How can this happen?
> 
> > +	get_avenrun(loadavg, FIXED_1/200, 0);
> > +	len = scnprintf(msg + len, *msglen - len, "load: %lu.%02lu  ",
> > +			LOAD_INT(loadavg[0]), LOAD_FRAC(loadavg[0]));
> > +
> > +	if (tty->ctrl.session == NULL) {
> > +		len += scnprintf(msg + len, *msglen - len,
> > +				 "not a controlling terminal\n");
> > +		goto out;
> > +	}
> > +
> > +	if (tty->ctrl.pgrp == NULL) {
> > +		len += scnprintf(msg + len, *msglen - len,
> > +				 "no foreground process group\n");
> > +		goto out;
> > +	}
> > +
> > +	p = pick_process(tty->ctrl.pgrp);
> 
> Why is no lock needed?
> 
> > +	if (p == NULL) {
> > +		len += scnprintf(msg + len, *msglen - len,
> > +				 "empty foreground process group\n");
> > +		goto out;
> > +	}
> > +
> > +	get_task_comm(tname, p);
> > +	getrusage(p, RUSAGE_BOTH, &rusage);
> > +	wallclock = ktime_get_ns() - p->start_time;
> > +
> > +	utime.tv_sec = rusage.ru_utime.tv_sec;
> > +	utime.tv_nsec = rusage.ru_utime.tv_usec * NSEC_PER_USEC;
> > +	stime.tv_sec = rusage.ru_stime.tv_sec;
> > +	stime.tv_nsec = rusage.ru_stime.tv_usec * NSEC_PER_USEC;
> > +	rtime = ns_to_timespec64(wallclock);
> > +
> > +	cputime = timespec64_to_ns(&utime) + timespec64_to_ns(&stime);
> > +	pcpu = div64_u64(cputime * 100, wallclock);

How is this number useful?

> > +
> > +	len += scnprintf(msg + len, *msglen - len,
> > +			 /* task, PID, task state */
> > +			 "cmd: %s %d [%s] "
> > +			 /* rtime,    utime,      stime,      %cpu,  rss */
> > +			 "%llu.%02lur %llu.%02luu %llu.%02lus %llu%% %luk\n",
> > +			 tname,	__get_pid(p, tty),
> > +			 (char *)get_task_state_name(p),
> > +			 rtime.tv_sec, frac_sec(rtime.tv_nsec),
> > +			 utime.tv_sec, frac_sec(utime.tv_nsec),
> > +			 stime.tv_sec, frac_sec(stime.tv_nsec),
> > +			 pcpu, getRSSk(p->mm));
> 
> Why do you think p->mm is still alive (even after the getRSSk()'s check)? So
> no get_task_mm() needed?
> 
> > +out:
> > +	*msglen = len;
> > +	return 0;
> > +}

Re lack of refcounting and locking, perhaps he's attemting a root hole?